APICS has been heavily involved with understanding and applying the theory of constraints (TOC) for a number of years now. Adopting and applying Lean principles has happened more recently—leading many people to wonder if TOC and Lean can and should work together.
The core tenants and rules associated with TOC can be found in Eli Goldratt’s best-selling book, The Goal, and a series of follow-up works. Of all the things Goldratt taught us, my favorite is the fundamental question: “Why are we in business? What is the goal?” I agree with Goldratt: It is to “make money now and in the future.”
The basic steps are:
- Identify the constraint.
- Exploit the constraint.
- Subordinate all other operations to exploit the constraint.
- If after numbers 2 and 3 more capacity is still needed to meet market demand, elevate the constraint.
- Go back to number 1: Don’t let inertia become the new constraint.
The effect of constraints can be visible internally (you can’t supply fast enough) or externally (there is not enough demand). Constraints can be tangible (a piece of machinery) or intangible (a policy). We must recognize that constraints often are intangible.
1. Identify the Constraint
It’s often pretty easy to find the constraint. In a manufacturing situation, you might find it simply by identifying where the biggest pile of work-in-process is in the plant. The equipment or process that has this pile in front of it, is probably the culprit—at least at the moment. In service businesses, it’s harder to see the constraints, but the same idea can apply. In an insurance and retirements benefits company I am working with, the constraints tend to move around with the business cycle, but certain areas do stand out; the large number of in-process claims waiting for specialists to review questionable claims is a clue.
This is where Lean tool knowledge and applications can be helpful, especially Value Stream Mapping (VSM) and work/process mapping observation tools. VSMs at the supply chain, enterprise, and process levels identify the presence or lack of flow and provide clues to the “what” and the “why.” My experience has been that properly developed VSMs identify all of the current and pending constraint points in a value stream, especially if the team designs the effort on the front end to collect the right kind of data.
Using a VSM approach at the macro supply chain level can be very helpful if the metrics therein include meaningful inputs that relate to volumes and capacity limits. I take this one step further in complex value streams by identifying the critical path. This is the series of activities that must happen with a start-to-start, start-to-finish, and finish-to-finish relationship.
For example, I recently applied this approach at a client in the business of designing and delivering standardized and custom software solutions for complex manufacturing and job shops. We tackled a key step in its business: the preparation and delivery of software demos. If you have been involved with creating the script for, and then delivering a demonstration of enterprise applications, you know this is extremely time consuming and costly, and it must involve a large number of subject matter experts from both the software company and the prospect.
Even though most of the key activities in the demo value stream happen concurrently, we were able to establish a critical path on a finish-to-finish basis and work backward from the delivery of a demo to the point the company authorizes the start of the process. Using this approach, we uncovered the constraints and a number of things that could be done to shorten the cycle, reduce the cost and improve the results at the same time. In just a few days we had an action plan underway that promised to cut costs up to 50%, decrease calendar time to delivery up to 50% and improve the win rate percentage, as well.
Standard work-study tools at a more granular level can be very revealing as well. For example, a work-study in the pulling module of a warehousing operation uncovered certain constraints caused by the practice of sorting pull tickets by box size before setting them up for pulling. The starter felt that by sorting the tickets by box size, efficiency would improve because workers could stand in one place in front of this stack and then make all the largest boxes first, medium boxes second and small boxes third for each store release and then send them down the conveyor for pulling.
While conventional wisdom might reward this idea, we learned that it actually caused a moving constraint. Why? Because large boxes have a lot more lines (and associated stock-keeping units) to pull, causing them to move much slower than medium or small boxes. These piles of large boxes were bogging down the pullers and causing a famine condition downstream. Plus, they created a new constraint—a bottleneck—in each pulling zone as they moved through the pull module.
The solution was simple: Stop sorting by box sizes. Instead, make and put up the boxes with their respective pull tickets as they come randomly and in standard sets as dictated by Takt time. While this did cause an apparent inefficiency for the starter, many other people could work more efficiently.
2. Exploit the Constraint
Getting the most out of the capacity constrained resource (CCR) is fairly easy. The team must make sure that utilization and efficiency are maximized at the CCR. This step is often indistinguishable from number 4, but it is usually beneficial to exploit before elevating a constraint. One immediate strategy is to build a supermarket buffer in front of the constraint. In this pull system, the buffer is enough work waiting to ensure the bottleneck never gets starved for work. If the supermarket is properly designed, the work pulled through is also likely to be the right things needed to keep customer satisfaction high and maximize the value of the systems outputs. Logically, the feeding operations should then be subordinated (the next step in TOC).
In addition, the CCR should immediately be examined by a waste reduction team to ensure you are getting the most possible out of it. If equipment is the constraint, go to relief strategies to keep it going during breaks, lunches and shift transitions. If people are the constraint, quickly identify the things they are doing that are not essential. For example, if claims specialists are spending time writing e-mails, this should be deferred to a lesser-skilled team member.
“Subordinate all other operations to exploit the constraint” now makes a lot of sense. As mentioned above, you must focus on high quality and reliable supply to the constraint and spend less effort worrying about utilization and efficiency at non-constraint resources. Skipping this step causes the majority of lost financial benefit, as anyone who has studied throughput accounting will tell you. (I encourage you to learn about throughput accounting. The best book I have seen on this is Throughput Accounting by Thomas Corbett.)
It’s equally important that you don’t neglect addressing culture issues. Be willing to make necessary changes to your measurements and culture. This is probably the most difficult step in TOC.
4. Elevate the Constraint
If steps 2 and 3 are not enough to meet current and expected future demand, increase capacity on the CCRs further. Add capacity through capital investment, outsourcing or offloading the constraint through alternative routings, process or product redesign, promotion of other product or services to the market and other measures.
Resist the temptation to increase investment in capital and people too soon. This is another area where Lean comes in. Charter teams to examine your CCR to improve results using these key tools:
- Total productive maintenance increases available time for production and increases the quality of outputs so that nothing produced is wasteful due to equipment capabilities.
- Employ setup or changeover reduction to decrease internal downtime on the CCR. This converts directly into increased throughput.
- Use Kaizen to improve flow, reduce movement and promote team focus on quality and productivity.
5. Don’t Let Inertia Become the New Constraint
It’s important to not let yesterday’s solutions become tomorrow’s new constraints. Policies, beliefs, and procedures that are not challenged can become barriers. What made sense today may not make sense tomorrow.
I suggest using zero-based improvement targets on CCRs that are continually identified and verified with the techniques suggested. New constraints identified may mean you must change policies—again!